Old Model
The old model of searching for root causes relied on checking the time when a Holdup/Idle state began and then checking another machine if it was not in the Failure/Lack of Components/Stopped by Operator state in the same period of time, regardless of the moment in which that state began.
...
On the other hand, it may also create a problem if a state is slightly misaligned in relation to a specific buffer. The farther from the source, the bigger the probability of “mistargeting” if the specified buffer time between machines is different from what it actually is (for example, that time can be greater due to slower work or shorter because of the bigger dimensions of products).
...
Improvements
In order to take into account full buffers / shorter reaction times, you take into consideration the entire period starting from the maximum reaction time (x) and ending in the moment when a stoppage began - or even a while after the stoppage started ( y ) which would let you spot a situation when a driver did not “report” a breakdown as fast as it should have:
...